Emergency Credit Can Save a Firm—and Still Leave Debt Behind
Government emergency credit for small businesses is meant to bridge the interval between a shock and a return of ordinary cash flow. This series examines the U.S. Small Business Administration’s disaster recovery loans and COVID-19 Economic Injury Disaster Loans (EIDL) to answer a practical question: when does borrowing restore a viable business, and when does it leave an obligation that outlasts the emergency? The answer requires more than asking whether a firm received money. It requires following what the money restored, what uncertainty it removed, and whether the business could later service the debt.
A loan can be valuable precisely because it is not a grant. Repayment gives public programs a way to recycle capital and can make them less costly than a transfer. But repayment also changes the risk borne by the borrower. If a damaged machine, building, or inventory position can be brought back into productive use, a loan may cover a temporary gap. If revenue will not return, the same loan adds another claim to cash flow that is already insufficient. The difference is not moral—whether the owner “used debt wisely”—but economic: whether credit restores earning capacity before its terms become restrictive.
The Cash Shortfall Arrives Before the Recovery
A disaster can damage a business in two connected ways. It can interrupt revenue by closing premises, reducing customer traffic, or disrupting suppliers. It can also damage the assets that make future revenue possible. A restaurant may lose equipment and fixtures; a manufacturer may lose machinery; a retailer may have damaged inventory or a premises it cannot reopen. Even a business that was viable before the shock may need cash before it can produce cash again.
That timing creates a financing problem. Private lenders must decide whether a business can repair, reopen, and repay while the local economy is still unsettled. The owner, meanwhile, faces bills and fixed obligations before the repair has generated new sales. Emergency credit is supposed to connect those two moments. In the favorable case, the loan pays for the repair that makes future income possible. The resulting income services the loan and allows the firm to resume relationships with other lenders, suppliers, landlords, and workers.
The unfavorable case has a different sequence. Credit arrives, but the underlying capacity does not recover: demand remains weak, damage is more expensive than expected, or the business was not viable after the shock. The loan can then postpone closure without changing its cause. A long maturity may make the monthly payment look manageable while extending the burden for years. The relevant distinction is therefore between a liquidity problem—a business that needs time and funds to get back to normal—and a solvency problem, in which the business cannot generate enough future value to support all its obligations.
The SBA’s pre-COVID disaster recovery program offers a focused setting for studying that distinction. It lends directly to businesses after declared disasters, primarily to repair or replace damaged property. In the research period from 2005 through 2017, 167,202 businesses applied and 54,532 borrowed. The program is not simply a pot of unrestricted cash: the SBA verifies losses, evaluates repayment capacity, and disburses funds in stages connected to the agreed recovery work. That design makes it possible to ask whether repair-oriented public credit changes a firm’s later path.
COVID-19 EIDL presents a related but different case. The program was expanded to address pandemic economic injury across a national disruption, not only damage to a particular physical asset. By the time it closed in 2022, it had approved about four million loans averaging approximately $96,000, and more than $300 billion in balances remained outstanding in late 2023. Unlike the Paycheck Protection Program, EIDL loans were not forgivable. The Federal Reserve report built from the 2023 Small Business Credit Survey finds that 44% of small employer firms had received an EIDL loan at some point: 28% still reported an outstanding balance and 16% had received one but paid it off.
Those figures do not by themselves tell two opposite stories—one in which disaster loans work and another in which EIDL failed. They describe different programs operating under different shocks and observed through different kinds of evidence. The natural-disaster study follows applicants over time and uses a credit-score threshold to estimate the causal effect of approval. The EIDL report compares firms by their debt status in a later survey. Its authors explicitly caution that firms still carrying EIDL debt may have been hit harder or had weaker prospects from the start. A high balance alongside financial stress is important, but it is not proof that borrowing caused the stress.
The diagram is a map of the question, not a verdict about either program. Both kinds of emergency credit can lead toward recovery or toward debt pressure. The distinction is what the loan changes in the business and what the evidence can establish about that change.
Two SBA Programs, Two Debt Stories
The natural-disaster recovery loans studied by the Census researchers are especially useful because they make the proposed bridge mechanism concrete. They are primarily intended to repair damaged property, and the SBA itself underwrites, originates, and services them. In the study period, the typical loan had a long maturity and a relatively low fixed rate, giving the borrower time to complete repairs and resume operations. The loan’s place in the capital structure also matters: the SBA claim is generally subordinate to existing secured debt but senior to secured debt taken afterward. If private lenders later extend credit, they do so behind a government claim that has helped make the business operational again.
That structure creates a testable sequence. If the loan merely replaces private money that would have been available anyway, private debt should fall or remain unchanged. If it only delays an inevitable failure, early survival might be followed by later distress. If it repairs a bottleneck that private lenders cannot confidently finance, the recipient should show not only greater survival but also improving debt service and renewed access to private capital. The research follows these outcomes instead of treating approval as success in itself.
EIDL has a different exposure to the bridge-versus-overhang problem. A 30-year loan at a modest interest rate can spread payments over a long period, which may be useful when a whole economy is disrupted. But the same horizon means that the balance can remain after the immediate emergency has faded. A firm that used the loan to cover a temporary revenue gap may recover and pay it down. A firm whose customers, labor supply, or business model changed permanently may carry the debt into a weaker future. The presence of a long repayment schedule does not answer which path occurred; it makes the path consequential.
The later survey provides signs of that consequence. Among firms with outstanding EIDL debt, 67% reported more than $100,000 in total outstanding debt, compared with 28% of firms that had never received EIDL. Fifty-two percent of the outstanding-balance group reported challenges making debt payments in 2023, compared with 29% of never-recipients. And 51% were operating at break-even or a loss at the end of 2022, compared with 39% of never-recipients. Firms with outstanding balances were also less likely to be fully approved when they applied for new financing, and debt was more likely to be cited as a reason for not applying.
These comparisons make the practical meaning of a lingering balance visible. Debt can constrain the next decision even when it did not immediately cause a firm to close: an owner may avoid new financing because taking on another obligation feels unsafe, or a lender may view the existing balance as evidence of limited repayment capacity. Yet the comparisons do not recreate the missing counterfactual. The firms that still owed EIDL may have been precisely those that suffered the largest losses, lacked insurance, or could not recover for reasons unrelated to the loan. Firms that repaid are not a clean control group either; repayment may reflect stronger businesses rather than an effect of the program.
This is why the two cases should be read together but not collapsed. The natural-disaster evidence asks, under a threshold-based comparison, what approval did to similar applicants near the cutoff. The EIDL evidence asks what firms reported about their financial condition while carrying, having repaid, or never taking on a balance. One provides unusually strong evidence about a causal effect in a defined population. The other documents a real debt burden and a borrower experience while leaving its causal origin unresolved.
The Question Is Not Whether to Borrow
The useful question is narrower and harder: what must be true for emergency credit to restore a business faster than it adds risk? Four tests organize the answer.
First, the loan must be connected to a capacity that can plausibly return. Physical repair is one example, but the general point is not that every crisis should be financed through repair loans. It is that the use of funds should have a credible relationship to future operating cash flow. A loan that only covers losses without changing the path to revenue may preserve time without creating repayment capacity.
Second, the credit should resolve a bottleneck that private markets cannot yet price. The natural-disaster study finds that recovery loans later increased private debt and other contracts rather than crowding them out. The researchers argue that the SBA’s repair process can reduce uncertainty about whether damaged assets will be restored. That mechanism is different from simply giving a firm more cash: it changes what other lenders can reasonably believe about the firm’s future.
Third, the repayment horizon must fit the recovery horizon. Deferral, fixed rates, and long maturities can prevent a payment schedule from arriving before repairs are complete. But patience is not forgiveness. If the business needs years of strong revenue to repay and the shock has permanently reduced demand, a low monthly payment can conceal rather than solve the mismatch.
Fourth, the evidence must be strong enough for the decision being made. The natural-disaster study finds persistent reductions in exit, bankruptcy, and delinquency and increases in employment, revenue, and private credit for the applicants affected by its approval thresholds. The EIDL report shows that outstanding debt is associated with financial pressure but cannot say whether EIDL caused it. Treating the latter as causal would overstate the evidence; treating it as irrelevant because it is not causal would ignore the lived constraint of repayment.
Emergency credit, then, is not automatically a bridge or automatically a burden. It is a bridge when it restores a viable source of cash flow, reduces uncertainty that blocks complementary finance, and gives repayment enough time to follow recovery. Debt lingers when the loan outlasts the business’s ability to recover, restricts the next financing decision, or is used where a grant, insurance payment, restructuring, or relocation decision would address the problem more directly. The next step is to examine how the natural-disaster study tests this sequence—and what happened to the firms that received the loan over the years that followed.